PHILPOTTS, Anthony R., Dept. of Geology and Geophysics, 
		The University of Connecticut, Storrs, CT, 06269

Dike rocks in the Eastern North American Mesozoic province provide 
the broadest geographic and geochemical sample of the associated 
igneous activity (Weigand and Ragland, 1970).  These rocks, 
however, are notoriously difficult to date accurately by radiometric 
methods.  The same is true of the associated lavas, but these at least 
can be given accurate stratigraphic ages.  Dating of the dikes, 
therefore, depends to a large extent on their geochemical correlation 
with a particular lava flow.
	Most of the dikes occur in the crystalline rocks surrounding 
the basins where their exposures provide samples of the magma that 
existed at some considerable depth in the crust and which may differ 
significantly from that of the associated lava.  In correlating dikes 
with lavas it is important, therefore, to consider what processes might 
cause differences between the magma at depth and on the surface.  
The enormous volumes of individual flood basalts and the great width 
of possible feeder dikes indicate that flow of magma was probably 
rapid and turbulent and, therefore, most unlikely to allow significant 
differentiation to occur during transit through the crust.  
Compositional differences between lava and feeder dike must 
consequently be due to changes in the composition of the magma 
rising from the source region or to processes acting during the 
solidification of the rocks.
	The Hartford Basin, like many of the other northern Mesozoic 
basins, contains three volcanic units which, from oldest to youngest, 
are the Talcott, Holyoke, and Hampden basalts.  Only three regional 
diabase dikes occur in southern New England that could possibly have 
fed these lavas.  Based on distinctive geochemistries, the Talcott 
Basalt has been correlated with the Higganum dike, the Holyoke 
Basalt with the Buttress dike, and the Hampden Basalt with the 
Bridgeport dike (Philpotts and Martello, 1986).  While the first and 
last of these correlations is almost perfect, the chemical variation in 
the Holyoke Basalt, which is extensive, does not overlap the 
compositional field of the Buttress dike samples, even though trace 
element ratios suggest they are related.
	Recent investigations of the Holyoke basalt (Philpotts et al., 
1996) indicate that where it contained a crystal mush that was in 
excess of 40 m thick, bulk density contrasts between the solids and the 
liquid resulted in compaction of the mush with expulsion of the 
interstitial liquid.  The entire range of composition of this basalt can 
be accounted for by this compaction, which is up to 28% where the 
flow is 174 m thick.  This same magma, solidifying in a vertical dike, 
should experience the same compaction forces as in the flow, except 
that they would be greater because of the greater vertical thickness of 
mush (kilometers rather than meters).  The Buttress dike composition 
can be interpreted as a Holyoke magma that has undergone 30 to 40% 
compaction during solidification.  Vertical stringers of pegmatitic 
segregation material in the central part of the Buttress dike are 
interpreted to be formed from the liquid expelled by compaction.  
These stringers have similar compositions to the coarse-grained 
segregation sheets in the Holyoke basalt.

Philpotts, A. R., and Martello, A., 1986,  Diabase feeder dikes for the
Mesozoic basalts in southern New England:  American Journal of
Science, v. 286, p. 105-126.
Philpotts, A. R., Carroll, M., and Hill, J., 1966,  Crystal-mush
compaction and the origin of pegmatitic segregation sheets in a thick flood
-basalt flow in the Mesozoic Hartford Basin, Connecticut:  Journal of
Petrology (in press--August Issue).
Weigand, P. W., and Ragland, P. C., 1970,  Geochemistry of Mesozoic
dolerite dikes from eastern North America: Contributions to Mineralogy
and Petrology, v. 29, p. 195-214.